Screening
Programme to detect unsuspected disease in a population of apparently healthy people
Surveillance
Programme to detect disease in a population already with disease
Important considerations
- Disease
- Common
- Important
- Long premobid latent period
- Detectable at early stage
- Treatable: by defined principles, cost-effective
- Test
- Sensitive (ability to detect)
- Specific (ability to exclude others)
- Non-invasive
- Acceptable to patients
- Cost-effective
- Does no significant harm to patients
Examples
- Breast cancer
- All women over 50-64 advised to have mammogram every 3 years.
- Mammographic abnormalities referred to breast specialist for clinical examination + further investigations
- 5-10% breast cancer familial
- Genetics: BrCA1 (chromosome 17), BrCA(chromosome 13), p53(chromosome 17), Ataxia telangectasia gene
- Detected cancers: smaller, CIS, well differentiated (ie. all rather good prognostic factors)
- Ovarian cancer
- Two or more 1st degree relatives
- BrCA1,BrCA2 genes
- Cervical cancer
- 3 yearly Papanicolau smears
- CIN 1,2,3
- Treated with cone excision biopsy
- Colorectal cancer: at risk families, polyps, IBD
- Faecal occult blood-testing kit, plus repeat test
- If positive > colonoscopy or double contrast barium enema
- Abdominal aortic aneurysms
- Congenital dislocation of hip
- Ortolani
- Barlow's
- Prenatal screening
Bias in screening
- Lead-time bias: Survival measured from detection to death will be longer (cause it's detected earlier)
- Selection bias: Individuals who take up screening are more health conscious
- Length bias: slowly growing tumours more likely to be detected by screening than rapidly growing tumours between screening intervals
Problems in screening
- Increased morbidity with unaffected prognosis
- Excessive therapy of doubtful cases
- Increased anxiety
- Lack of target population co-operation
- Costs
- Inffective screening tests
- False reassurance